Bringing back exams

Our Middle School is about to reintroduce semester exams in the academic subjects such as Maths, English, HASS, Science and Languages. There are many arguments for and against having exams in a Middle School environment. I have changed my mind on this form of assessment and my reasons are not because I hate young people.

A few years ago, I argued for removing all exams in our Middle school environment. I didn’t see the point of them because they were not assessing what I thought was most important, critical and creative thinking skills. The arguments placed forward to keep them were all based around preparing students for what was coming in Year 12. Our loudest exam advocates wanted students to be familiar with the exam environment so the ‘real thing’ would be less stressful. I thought this was a pretty weak argument, I still do.

However, I am now convinced of the necessity of a broad, accumulated domain-specific knowledge stored in long-term memory. Without relevant knowledge easily accessible in long-term memory, the chance of being able to critically and creatively think about anything is little to none. I’ve written in more detail about this here.

So why do I want exams? Put simply, exams will encourage students to more frequently revisit previously learnt content. Cognitive scientists suggest that requiring the brain to recall information previously learnt at regular intervals is one of the best bets for knowledge to stick. Go here and here for more details.  I am not suggesting cramming for an exam overnight will help students. Rather, if students are required to revisit material they have learnt over 4-5 months in a low-stakes environment, they are more likely going to keep that information accessible in long-term memory for a longer period of time. Especially if exams are complemented by frequent low-stake quizzes throughout a semester.

If students have a broad, accumulated domain-specific knowledge stored in long-term memory, they are going to perform much better at tasks which require critical and creative thinking. For example, in History, wouldn’t it be extremely beneficial if students could still remember content they had learnt about WW1 (Year 9) in Year 12?  When students are required to analyse and evaluate the significance of one event, individual or source, they can draw upon knowledge regarding the details of a multitude of other time periods. Imagine if students could remember the themes, ideas and literary techniques of the many novels they studied in previous English classes. This knowledge would assist them tremendously when they need to critique another author’s literary style. This idealistic dream has a much better chance of happening if students are required to revise accumulated information throughout secondary school via regular low stakes exams.

Often people who argue against learning a lot of content in modern education claim that it will be forgotten fairly quickly, so why bother. This is true if what is taught once is never revisited. However, if content which was learned in previous years was regularly included in relevant quizzes and exams, forgetting might become a little bit harder.

Do I think exams are the best form of assessment? Definitely not. But from this paradigm, an exam’s chief purpose is focused on the revision it requires. Any assessment value is a bonus and this comes from the measurement of what has been retained over the long term.

Novices use thinking skills. Experts use knowledge.

Until fairly recently, I thought the Australian education system needed a total upheaval. I believed a call for revolution. The message goes something like this: with the introduction of the internet, the entire world’s knowledge is at our fingertips. Anyone can become an expert about anything. If the content is cheap, then the real value is in one’s thinking. Schooling across the twentieth century was all about giving students knowledge that could only be found in books. Now because every kid has a smartphone with access to Wikipedia, 20th-century education is redundant. Along with the fax machine, USBs and taxis. The new goal is thinking. Deep thinking.

Apparently, because modern-day students have access to Google, we should focus on teaching them how to ‘think deeply’ about the information they find. No need to waste time filling young people’s heads with information. The internet can store the information for them. In the twenty-first century, students can quickly look up relevant information about any topic they please and then focus on the critical and creative thinking. This is the new money maker.

Every conference I went to, from EduTech to the Google Summit was calling for teachers and school leaders to tear down traditional forms of schooling and erect in its place, a student-driven approach, as this would encourage ‘the deep learning’. However, I kept stumbling on one of John Hattie’s effects, labelled ‘student control over learning,’ which is placed as having a 0.01 effect size. If a teacher did nothing else but smile at a young person every day, it would probably have more of an effect than that pitiful value.

Then I became aware of some inconvenient truth. Inconvenient, because it has caused me to doubt everything I have have been doing thus far in my teaching career. Even more so, because it contradicted everything I had been banging on about to the staff at my school.

Greg Ashman (check out his fantastic blog here) introduced me to Cognitive Load Theory. To be honest, it all seemed too complicated for my simple mind until one day I read this. And it convinced me. I realised that the evangelists calling for this revolution have totally ignored the current research regarding human cognitive architecture. This is why things just didn’t add up.

The brain does not work as conveniently as the revolutionaries would have us believe. Cognitive Load Theory highlights the differences between how information is processed in working memory and long-term memory. Our working memory can retain novel information while simultaneously processing other information; however, there are significant limitations. It can only hold between three and four individual items at one time (Cowan, 2001). The duration is also limited, with novel information being held in working memory approximately no more than a few seconds. It’s why when someone rattles off their phone number too quickly when leaving you a message, you may need to relisten to the message multiple times to finally get the 10 digits on paper. Our conscious minds can only hold a few novel items at once, and only for only a brief moment. But why can you easily recall your own 10 digit phone number? Because this is stored in your long-term memory, and when information is processed from long-term memory, there are no known limits on either capacity or duration (Sweller et al, 2011). This is a big deal.

A study on chess grandmasters illustrates this important difference between working memory and long-term memory. De Groot (1965) took a series of Chess grandmasters and faced them against a group of weekend players. Instead of playing chess, he showed them a series of chess configurations for five seconds. He then asked each group to recall the positions of the chess pieces. The grandmasters had a vastly superior recall ability compared with the casual players. However, when the chess pieces were spread across the board at random, the grandmasters lost their edge showing the same limitations of working memory as the recreational players. De Groot’s study, alongside various replications from Chase & Simon (1973), disproved an intuitive but common misconception. They found grandmasters were not actually expert problem-solvers or strategists, as they shared very similar thinking routines with casual players when making decisions. What made the biggest difference was grandmasters had memorised thousands of board configurations and possible subsequent moves into their long-term memory, which could be quickly recalled. The findings of this study have been replicated many times in varying domains (Ericsson & Kintsch, 1995). So how does this relate to thinking and learning?

“Novices need to use thinking skills. Experts use knowledge” (Sweller, Ayres & Kalyuga, 2011, p. 21).

When examining the cognitive architecture, we find that, ironically, the more domain knowledge we have in long-term memory, the more we can actually use critical and creative thinking skills. If we rely on using problem-solving skills without being able to draw upon relevant domain-specific knowledge, we are at a significant disadvantage (Kirschner, Sweller & Clark, 2006).

The following video is of John Sweller, a champion of Cognitive Load Theory giving a fascinating lecture elaborating on the essential elements of this theory:

The internet is an immeasurably useful tool for teaching and learning. However, because of our limited working memory, students still need the time to patiently learn a vast amount of subject knowledge. We cannot simply tap into the internet and use our magical thinking skills to become overnight experts. The hard truth is learning takes time, and to have any thoughts of value, let alone originality, people usually need to know a large amount of relevant content knowledge first. Consequently, schools should not withhold subject content from their students to excessively emphasise teaching thinking skills.

Instead of continually dismissing subject content, the loudest voices influencing Australian education ought to focus on how schools can most effectively teach students to transfer subject knowledge into their long-term memory. Counterintuitively, this is actually our quickest and most likely path to empowering our students to become life-long critical and creative thinkers.

Reference List: 

Chase, W., & Simon, H. (1973). Perception in chess. Cognitive Psychology, 4(1), 55-81.

Cowan, N. (2001). The magical number 4 in short-term memory: A reconsideration of mental storage capacity. Behavioral & Brain Sciences, 24(1), 87-185.

De Groot, A. (1965). Thought and choice in chess. The Hague: Mouton.

Ericsson, K. A., & Kintsch, W. (1995). Long-Term Working Memory. Psychological Review, 102(2), 211-45.

Kirschner, P., Sweller, J., & Clark, R. (2006). Why Minimal Guidance During Instruction Does Not Work: An Analysis of the Failure of Constructivist, Discovery, Problem-Based, Experiential, and Inquiry-Based Teaching. Educational Psychologist, 41(2), 75-86.

Sweller, J., Ayres, P., & Kalyuga, S. (2011). Cognitive load theory. New York: Springer.